Dataset Information

Leveraging Machine Learning for Advanced Nanoscale X-ray Analysis: Unmixing Multicomponent Signals and Enhancing Chemical Quantification.

ABSTRACT: Energy dispersive X-ray (EDX) spectroscopy in the transmission electron microscope is a key tool for nanomaterials analysis, providing a direct link between spatial and chemical information. However, using it for precisely determining chemical compositions presents challenges of noisy data from low X-ray yields and mixed signals from phases that overlap along the electron beam trajectory. Here, we introduce a novel method, non-negative matrix factorization based pan-sharpening (PSNMF), to address these limitations. Leveraging the Poisson nature of EDX spectral noise and binning operations, PSNMF retrieves high-quality phase spectral and spatial signatures via consecutive factorizations. After validating PSNMF with synthetic data sets of different noise levels, we illustrate its effectiveness on two distinct experimental cases: a nanomineralogical lamella, and supported catalytic nanoparticles. Not only does PSNMF obtain accurate phase signatures, but data sets reconstructed from the outputs have demonstrably lower noise and better fidelity than from the benchmark denoising method of principle component analysis.

SUBMITTER: Chen H

PROVIDER: S-EPMC11342375 | biostudies-literature | 2024 Aug

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Leveraging Machine Learning for Advanced Nanoscale X-ray Analysis: Unmixing Multicomponent Signals and Enhancing Chemical Quantification.

Chen Hui H Alexander Duncan T L DTL Hébert Cécile C

Nano letters 20240806 33

Energy dispersive X-ray (EDX) spectroscopy in the transmission electron microscope is a key tool for nanomaterials analysis, providing a direct link between spatial and chemical information. However, using it for precisely determining chemical compositions presents challenges of noisy data from low X-ray yields and mixed signals from phases that overlap along the electron beam trajectory. Here, we introduce a novel method, non-negative matrix factorization based pan-sharpening (PSNMF), to addres ...[more]

PMID: 39106344

Similar Datasets

Project description:Population expansion and rising consumer demand for nutrient-dense meals have both contributed to an increase in the consumption of animal protein worldwide. A significant portion of the meat and eggs used for human consumption come from the poultry industry. Early diagnosis and warning of infectious illnesses in poultry are crucial for enhancing animal welfare and minimizing losses in the breeding and production systems for poultry. On the other hand, insufficient techniques for early diagnosis as well as infectious disease control in poultry farms occasionally fail to stop declining productivity and even widespread death. Individual physiological, physical, and behavioral symptoms in poultry, such as fever-induced increases in body temperature, abnormal vocalization due to respiratory conditions, and abnormal behavior due to pathogenic infections, frequently represent the health status of the animal. When birds have respiratory problems, they make strange noises like coughing and snoring. The work is geared towards compiling a dataset of chickens that were both healthy and unhealthy. 100 day-old poultry birds were purchased and split into two groups at the experimental site, the poultry research farm at Bowen University. For respiratory illnesses, the first group received treatment, whereas the second group did not. After that, the birds were separated and caged in a monitored environment. To eliminate extraneous sounds and background noise that might affect the analysis, microphones were set a reasonable distance away from the birds. The data was gathered using 24-bit samples at 96 kHz. For 65 days, three times per day (morning, afternoon, and night) of audio data were continually collected. Food and water are constantly provided to the birds during this time. During this time, the birds have constant access to food and water. After 30 days, the untreated group started to sound sick with respiratory issues. This information was also noted as being unhealthy. Chickens' audio signals were recorded, saved in MA4, and afterwards converted to WAV format. This dataset's creation is intended to aid in the design of smart technologies capable of early detection and monitoring of the status of birds in poultry farms in a continuous, noninvasive, and automated way.

Dataset Information

Leveraging Machine Learning for Advanced Nanoscale X-ray Analysis: Unmixing Multicomponent Signals and Enhancing Chemical Quantification.

Publications

Leveraging Machine Learning for Advanced Nanoscale X-ray Analysis: Unmixing Multicomponent Signals and Enhancing Chemical Quantification.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets